Compared with conventional bare-metal stents (BMS), the introduction of drug-eluting stents (DES) has resulted in a substantial reduction in the incidence of in-stent restenosis.1 DES systems eluting either sirolimus2 or paclitaxel3 from a polymer stent coating have been shown in randomised trials to effectively inhibit the process of neointimal proliferation, resulting in restenosis reduction.
However, despite these advances, restenosis rates remain substantial in patients at high risk of restenosis, e.g. patients with long lesions and complex lesion morphologies or patients with lesions in bifurcations.4 Furthermore, the incidence of late-occurring thrombotic vessel occlusions5,6 and the development of late restenosis7 have raised issues about the safety and efficacy of DES in the long term. Both of these late-occurring complications have been related to a marked inflammatory response against the non-degradable polymer-coated stent surface, as well as to an incomplete re-endothelialisation.
In the Individualized drug-eluting Stent system to Abrogate Restenosis (ISAR) project, we developed a novel DES system that allows for an individualisable, drug- and dose-adjustable stent coating without the need to use polymers. The system consists of two components: a balloon-mounted stainless steel stent with a unique microporous strut surface and a coating device that can be operated in the catheterisation laboratory. The purpose of the ISAR study was to evaluate the feasibility, safety and efficacy of this novel DES system. For the proofof- principle of this concept, stents were coated with increasing doses of rapamycin, a cytostatic compound that has been proved to inhibit neointima formation.
Methods
Study Design
ISAR was a prospective, open-label, dose-finding study for the evaluation of four sequentially increasing rapamycin doses in a polymer-free stent coating for the prevention of restenosis. Patients were considered eligible for this study if they complained of angina pectoris or had exercise-induced ischaemia in the presence of angiographically significant stenosis in native coronary arteries with a reference-vessel diameter between 2.5 and 3.5mm. Patients suffering from an acute myocardial infarction (MI) (within the last 72 hours before the intervention), those with lesions in the left main stem and those with lesions with in-stent restenosis were excluded.
The placement of multiple rapamycin-eluting stents at the same dosage was allowed to cover one or more lesions. The control group consisted of patients treated with the same microporous BMS without a drug coating. The protocol was approved by the institutional ethics committee and all patients gave written, informed consent.
The Drug-eluting Stent Platform
The DES platform consists of two components: a pre-mounted, micro-structured 316l stainless steel microporous stent in a disposable coating cartridge and the coating device. The detailed process of stent coating has been described elsewhere.8 The microporous stent surface increases the drug reservoir capacity and allows for a retarded drug release without the need to apply a polymer. For stent coating, the cartridge holding the stent system is placed into the coating device and a 1ml drug reservoir containing the dissolved drug in a pre-defined volume is connected to the cartridge. The coating process is initiated by advancing the drug into a mobile, positionable ring containing three jet units, which allow for uniform delivery of the drug onto the stent surface. After the spray coating, the stent surface is dried by removing the solvent with pressured air. The stent is then available for immediate use.
The stents in this study were coated with 0.5, 1.0 and 2.0% rapamycin solutions. Specific analyses with high-performance liquid chromography (HPLC) have shown that the spray coating with these rapamycin solutions resulted in 138±14, 313±59 and 479±26μg rapamycin/mm2 stent surface area, respectively. Furthermore, pharmacokinetic assays revealed that rapamycin is released for more than three weeks, with more than two-thirds released within the first week. The available stent sizes were 8, 12, 16, 20, 23 and 25mm, with diameters of 2.0, 2.5, 3.0 and 3.5mm.
Results and Clinical Outcome
In total, 602 patients were sequentially assigned to receive a microporous BMS (155 patients) and a rapamycin-eluting stent with a dose of 0.5% rapamycin (139 patients), 1.0% rapamycin (161 patients) and 2.0% rapamycin (147 patients). Clinical characteristics were comparable between groups.
Notably, 29% of patients had diabetes mellitus and more than one-third of patients had already suffered from MI. A total of 708 coronary stenoses were treated in this study, with a higher prevalence of stenoses in the left anterior descending (LAD) in the 1% rapamycin group. Additional differences between the sequential dose groups were seen for the incidence of complex lesions, chronic occlusions and vessel size. No problems were encountered during on-site stent coating, and the placement of coated stents was successful in all patients.
None of the patients died during the first 30 days after stent placement. Two patients (0.3%) out of the overall study population suffered from a thrombotic stent occlusion: one lesion each in the 0.5% and 2.0% rapamycin groups. Both occlusions occurred within 24 hours after rapamycin-eluting stent placement. The complication of non-fatal MI within the first month after stenting was observed in 1.3% of the BMS patients and in 0.7, 2.5 and 2.0% of the 0.5, 1.0 and 2.0% rapamycin-eluting stent patients, respectively (p=0.64). At one-year follow-up, the combined rate of death or MI did not differ between treatment groups. The incidence of death or non-fatal MI at one year was 3.9% in the BMS patients and 1.4, 3.7 and 2.7% among the 0.5, 1.0 and 2.0% rapamycin-eluting stent patients, respectively (p=0.59).
Incidence of Angiographic and Clinical Restenosis
Repeat angiography was performed in 484 of 598 eligible patients (80.9%) at a median time interval of 198 days (inter-quartile range: 175–213 days). A total of five patients died before the scheduled angiographic follow-up; the remaining patients declined to undergo the procedure. Compared with the BMS group, the in-stent restenosis rate demonstrated a significant dose-dependent reduction with increasing rapamycin doses (p=0.012). Similarly, the in-segment restenosis rate was reduced with increasing rapamycin doses (p=0.024). Consequently, the in-segment late lumen loss was lowest – at 0.36±0.55mm – in the 2.0% rapamycin-eluting stent group. Target lesion revascularisation (TLR) was necessary in 40 BMS lesions (21.5%) and in 28 (16.4%), 23 (12.6%) and 15 (8.8%) of the 0.5, 1.0 and 2.0% rapamycin-eluting stent patients, respectively (p=0.006). Thus, the need for TLR was significantly reduced – by 59% – in the 2.0% rapamycin-eluting stent group (see Figure 1).
We applied a multivariable logistic regression model to correct for the possible influence of the differences in baseline patient and lesion characteristics on in-segment restenosis and on TLR.
In this analysis, the well-known variables of presence of diabetes, long lesions and small vessel size – among others – were identified as independent predictive risk factors for restenosis. In addition, the placement of rapamycin-eluting stents with increasing drug doses was independently associated with a reduced risk of restenosis (p=0.005) and TLR (p<0.001).
Discussion
In this prospective, sequential, dose-finding study, a high proportion of patients presenting with acute coronary syndromes, multivessel disease and complex lesions was included. The study demonstrates that the placement of a polymer-free rapamycin-eluting stent with a microporous strut surface was feasible and safe. Compared with BMS, rapamycin-eluting stents markedly reduced the risk of restenosis in patients with a wide range of coronary lesions. All indices of restenosis improved with increasing rapamycin doses on the stent. Compared with the BMS group, the biologic potency of the polymer-free rapamycin-eluting stent was demonstrated by a maximal 43% relative reduction in the risk of angiographic in-segment restenosis, with a corresponding 59% reduction in the need for TLR.
Since the introduction of DES in interventional cardiology, several large clinical trials have consistently demonstrated an impressive reduction of in-stent restenosis in de novo coronary lesions by rapamycin- and paclitaxel-eluting stents.1,3,9,10 However, restenosis rates remain substantial in patients at high risk of restenosis, e.g. those with challenging interventional scenarios such as long lesions with complex morphologies, lesions in bifurcations4 or in-stent restenotic lesions. To overcome these limitations, dose adjustments may be desirable to enable an individual dosage for specific lesion or patient subsets. In addition, the currently approved DES platforms use a polymer-based coating for retardation of drug release. Polymer-based DES have been associated with prolonged inflammatory reactions,5,11 and several investigators believe that the polymer is the most likely mechanism.5 This hypersensitivity reaction may cause late stent thrombosis and restenosis.6,7 Therefore, we developed a novel concept for a DES platform that allows for dose-adjustable coating of stents without the need to use a polymer.
This DES platform consists of two components: a microporous stainless steel stent and a coating device that can be operated in the catheterisation laboratory. Compared with a non-microporous stent, the microporosity may reduce neointima proliferation through an accelerated re-endothelialisation,12 allowing for an increase in the drug reservoir and a retardation of drug release. Indeed, pre-clinical investigations demonstrated that increasing doses of a chosen drug could be sprayed on a microporous stent surface and that the potential loss of the drug in the transition between insertion of the stent system into the guiding catheter and stent placement is negligible.8 Furthermore, studies in the porcine restenosis model demonstrated that the use of the microporous stent system is associated with a significant reduction in neointima proliferation when rapamycin is used as the active antiproliferative compound.8
So far, an optimal release kinetic has not been determined for any compound used on DES that is able to obtain a balance between the inhibition of neointima proliferation and re-endothelialisation of the injured coronary artery. However, in a small study population the slow-and fast-release kinetics of rapamycin from the polymer-based stent system have resulted in similar efficacies in inhibition of neointima growth.9 In addition, short-term oral treatment with rapamycin has been shown to be effective for restenosis prevention in patients with recurrent in-stent restenosis.13 These results indicate that a brief treatment duration may be sufficient to inhibit restenosis formation when rapamycin is the administered compound. With the current stent system, two-thirds of the stent-based rapamycin is eluted within the first week and the rest in the next two weeks, resulting in an effective inhibition of neointima proliferation. Although an even stronger inhibition may be obtainable with a retardation of rapamycin release, the achievement of the current results without the use of a polymer may have advantages in the event-free long-term follow-up of patients treated with rapamycin-eluting stents.
The efficacy of non-polymer-based DES has been questioned by the results of previous randomised trials. Although the placement of a non-polymer-based paclitaxel-coated stent has resulted in an effective, dose-dependent reduction of in-stent neointimal tissue proliferation,14 a similar non-polymer-based paclitaxel coating was associated with only a non-significant reduction of restenosis in the DELIVER trial.15 The differences in restenosis prevention between the non-polymer-based paclitaxel-coated stents and our polymer-free rapamycin-eluting stent may be explained by the differences in the release kinetics of the drugs provided by the different strut surface and the drugs themselves.
Limitations
The groups of patients were enrolled in this study in a sequential manner; thus, we acknowledge the limitations of a non-randomised trial. Although the current dose-finding study shows that the on-site drug coating of stents is feasible and safe, and that a dose response is obtainable with increasing drug doses on the stent surface, it did not investigate comprehensively the potential of the system in terms of an individualised dosing regimen for different lesion morphologies or the use of different drug combinations on the stent system. Two of 447 patients (0.4%) receiving a rapamycin-eluting stent suffered from stent thrombosis within the first 24 hours after stent placement. This rate of thrombotic stent occlusions is comparable to that found in other DES trials.2,3 Nevertheless, we acknowledge that the sample size of this study is too small for a comprehensive investigation of the impact of rapamycin-eluting stents on events such as stent thrombosis.
Current DES are an expensive treatment option with the potential to reduce medical care costs during follow-up.16 Although the rapamycin-eluting stent assessed in the ISAR study may be a less expensive option, no formal cost-effectiveness analysis was performed.
Acknowledgements
The ISAR project was supported by a research grant (AZ 504/02) from the Bayerische Forschungsstiftung, Munich, Germany.
This article was originally published in its entirety in European Heart Journal, 2005;26:1475–81 (doi:10.1093/eurheartj/ehi405), and is reprinted with permission.